[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M698QSJ3BTTMXBFCB7TG5D7S":3,"skill-same:01M698QSJ3BTTMXBFCB7TG5D7S":64},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":36,"linkOnly":37,"markdownContent":38,"markdownBody":39,"frontMatter":40,"refName":41,"commitSha":42,"skillDirectory":43,"forksCount":44,"repositoryHtmlUrl":45,"repositoryHomepageUrl":46,"ownerHtmlUrl":47,"ownerBlogUrl":46,"ownerDescription":48,"updatedAt":49,"install":50,"resources":63},"01M698QSJ3BTTMXBFCB7TG5D7S","windows-pr-prepare",null,"Use before opening a Windows node PR, writing its description or updating its body. Prepare a local draft with validation and review evidence; publish only with explicit authority.","openclaw\u002Fopenclaw-windows-node","openclaw","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F252820863?v=4","01M698P3FSAZ6B4R928ZF19KPT","01M698Q6SND6RHZDPP8ZM0NSP7",2198,"MIT","https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fwindows-pr-prepare\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35],"windows pr prepare","windows","prepare","pr","pr prepare","body","writing","authority","with","local","draft","evidence","only","use","node","publish","validation","review","2026-11-10 10:12:30",false,"---\nname: windows-pr-prepare\ndescription: \"Use before opening a Windows node PR, writing its description or updating its body. Prepare a local draft with validation and review evidence; publish only with explicit authority.\"\n---\n\n# Windows PR preparation\n\nUse before opening a PR, writing its description or updating its body. Follow\nthis flow for the selected change, not a repository-wide audit. The preparation\noutput is a local body and a ready-or-blocked handoff. Publication and merge\nrequire their own authorization.\n\n## 1. Establish the claim, candidate and authority\n\nRead [repository instructions](..\u002F..\u002F..\u002FAGENTS.md), the affected production owner\nand existing tests. State the concrete behavior being changed and the invariant\nto preserve. For instructions, identify the reader's task and authoritative\nguidance instead of inventing a product behavior claim.\n\nRecord the repository, task worktree, intended target, observed target tip,\ncomparison base, HEAD and relevant staged, unstaged and untracked content.\nFor uncommitted work, retain the test or proof runner's record of the source it\nused. If there is no such record, capture the diff and relevant file hashes when\nrunning the check and label that association as manually recorded. Do not infer\nan old run's source from today's files or require a commit just to obtain a SHA.\nMissing comparison context stays explicit.\n\nSeparate permission to inspect, implement and validate from permission to install,\nchange shared runtimes, commit, push, upload, edit GitHub metadata or merge.\nDo not fetch, rebase or switch branches merely to make evidence identities agree.\n\n## 2. Select the smallest sufficient validation\n\nUse the [validation planner](..\u002F..\u002F..\u002Fdocs\u002FTEST_COVERAGE.md#explain-the-validation-plan)\nwith all candidate changes, then the [proof skill](..\u002Fopenclaw-proof-validation\u002FSKILL.md)\nand [proof pools](..\u002F..\u002F..\u002Fdocs\u002FPROOF_POOLS.md#declare-pools-in-a-pr) for applicable\nobligations. A planned lane or declared pool is not an executed result.\n\nOrdinary prose uses [documentation and skill checks](..\u002Ftechnical-documentation\u002FSKILL.md),\nnot full-product builds or native runtime proof. Code retains mandatory local\ncloseout plus focused checks; the shipped `winnode` reference retains its command\ncontract checks.\n\nSelect a runtime only when the claim needs one and authority permits it. A\nseparate app profile does not authorize installing or registering a Gateway or\nchanging a shared runtime. For MCP, follow the\n[local MCP guide](..\u002Fwindows-app-mcp\u002FSKILL.md), including fresh-process\nprofile\u002Ftoken\u002FURL binding.\n\n## 3. Check what actually ran\n\nApply the existing\n[result rules](..\u002F..\u002F..\u002Fdocs\u002FTEST_COVERAGE.md#accepting-result-reports).\nFor inventory-bound evidence, use the\n[validator and limits](..\u002F..\u002F..\u002Fdocs\u002FPROOF_POOLS.md#validate-local-execution-evidence).\n\n- Read outputs from the exact invocation, not the newest file or exit zero alone.\n  Check executed cases, pass\u002Ffail\u002Fskip counts and failures. Label manual\n  observations as manual; do not fabricate receipts.\n- Distinguish the PR head, tested commit (including CI merge commits), uncommitted\n  source used by the run, and loaded binary\u002Fruntime.\n- After edits, commits or rebase, reassess which results and reviews still apply.\n  Source equality can establish correspondence, but cannot turn an earlier\n  review or runtime execution into a new run.\n- Retain original failures. An instrumented pass does not erase them, and\n  restoring source does not restore previously built binaries.\n- Ordinary allowed skips are not complete custom\u002Fnative proof. Identify the CI\n  run and attempt when citing results; do not present older results as newly\n  executed tests. Successful dispatch is not completed testing or review approval.\n\nMissing required review, UI or Gateway evidence stays `Not verified \u002F blocked`.\nFor an unexpected failure, preserve the attempt and identify the next bounded\ninvestigation or decision. Fix task-caused defects within scope; do not\nautomatically repair unrelated tests, rerun until green or create new PR layers.\nA failure in another task blocks this task only if this task depends on that\nresult. Otherwise continue the authorized work without borrowing its evidence.\n\n## 4. Review the intended scope\n\nAccount for both source review and the required independent rubber-duck review:\n\n- For source review, when requested or required, use the\n  [bounded local-review recipe](..\u002F..\u002F..\u002Fdocs\u002FAGENT_WORKFLOWS.md#run-a-scoped-shared-review).\n  Select mode\u002Fbase explicitly, include committed and dirty work when intended,\n  and retain Windows P0-P3 coverage. Follow its context, final-output and fallback\n  rules.\n- Before publishing non-trivial UI, MCP, node-command, setup, pairing, security,\n  permissions or diagnostics changes, obtain an independent rubber-duck review\n  of the implementation and proof plan. Follow the\n  [proof skill's review route](..\u002Fopenclaw-proof-validation\u002FSKILL.md#collect-the-selected-proof)\n  for available roles and explicit blockers. A source-review verdict alone does\n  not establish that this review was performed.\n\nReport actual scope, findings, dispositions and missing review. Self-checks are\nnot independent review, and an unavailable required review remains blocked.\nVerify findings before fixing, then refresh affected checks and review coverage.\nReviews are advisory, not runtime proof or authority to publish.\n\n## 5. Write the local body from the actual template\n\nRead the current [Windows PR template](..\u002F..\u002F..\u002F.github\u002Fpull_request_template.md).\nCopy it to a local session artifact; preserve headings, HTML comments, checklist\nitems and ordering, including security, compatibility and review conversations.\nFollow its title convention. Check only supported claims, not aspirational work.\n\nThe [proof package](..\u002Fopenclaw-proof-validation\u002FSKILL.md#pr-proof-package)\ndefines what counts as evidence. This step assembles that evidence into the\ncurrent template:\n\n- Lead with the current behavior, strongest evidence and meaningful limits.\n  Keep risks and blockers visible.\n- Put exact commands and actual outcomes\u002Fcounts in Validation. Identify the\n  tested source in Real Behavior Proof, including uncommitted work.\n- For prose-only work, label source\u002Flink checks as static assessment and runtime\n  proof as not applicable.\n- Declare every applicable pool, or use ``- `none`: \u003Creason>`` when no custom\n  pool is required.\n\nKeep supporting history in links or optional details, not an operational diary.\nCompare the completed body with the template; do not drop unanswered fields or\nclaim pending checks passed. Apply the\n[evidence\u002Fredaction boundaries](..\u002F..\u002F..\u002Fdocs\u002FPROOF_POOLS.md#evidence-and-secret-boundaries)\nand verify referenced artifacts. Private local paths are not public proof links.\n\n## 6. Hand back readiness and stop\n\nReport the exact candidate, local draft location, strongest evidence and remaining\nlimits. Say **ready for human review** only when applicable gates are satisfied.\nMissing required source or rubber-duck review means **blocked**, even when tests\npass. Name any other **blocked**, **waiting for required CI**, or **candidate\nchanged** state and the next bounded decision. A complete local draft can still\ndescribe blocked work; it is not a ready-to-merge claim.\n\nStop before publication without explicit authority. If separately authorized,\nverify the intended repository\u002Fbase\u002Fhead and live PR state, then use the host's\nsupported publication tools and their guard\u002Ffailure contracts.\nPreserve concurrent human edits; after an uncertain write, inspect the result\nbefore retrying.\nVerify published content and links without treating publication as proof.\n","# Windows PR preparation\n\nUse before opening a PR, writing its description or updating its body. Follow\nthis flow for the selected change, not a repository-wide audit. The preparation\noutput is a local body and a ready-or-blocked handoff. Publication and merge\nrequire their own authorization.\n\n## 1. Establish the claim, candidate and authority\n\nRead [repository instructions](..\u002F..\u002F..\u002FAGENTS.md), the affected production owner\nand existing tests. State the concrete behavior being changed and the invariant\nto preserve. For instructions, identify the reader's task and authoritative\nguidance instead of inventing a product behavior claim.\n\nRecord the repository, task worktree, intended target, observed target tip,\ncomparison base, HEAD and relevant staged, unstaged and untracked content.\nFor uncommitted work, retain the test or proof runner's record of the source it\nused. If there is no such record, capture the diff and relevant file hashes when\nrunning the check and label that association as manually recorded. Do not infer\nan old run's source from today's files or require a commit just to obtain a SHA.\nMissing comparison context stays explicit.\n\nSeparate permission to inspect, implement and validate from permission to install,\nchange shared runtimes, commit, push, upload, edit GitHub metadata or merge.\nDo not fetch, rebase or switch branches merely to make evidence identities agree.\n\n## 2. Select the smallest sufficient validation\n\nUse the [validation planner](..\u002F..\u002F..\u002Fdocs\u002FTEST_COVERAGE.md#explain-the-validation-plan)\nwith all candidate changes, then the [proof skill](..\u002Fopenclaw-proof-validation\u002FSKILL.md)\nand [proof pools](..\u002F..\u002F..\u002Fdocs\u002FPROOF_POOLS.md#declare-pools-in-a-pr) for applicable\nobligations. A planned lane or declared pool is not an executed result.\n\nOrdinary prose uses [documentation and skill checks](..\u002Ftechnical-documentation\u002FSKILL.md),\nnot full-product builds or native runtime proof. Code retains mandatory local\ncloseout plus focused checks; the shipped `winnode` reference retains its command\ncontract checks.\n\nSelect a runtime only when the claim needs one and authority permits it. A\nseparate app profile does not authorize installing or registering a Gateway or\nchanging a shared runtime. For MCP, follow the\n[local MCP guide](..\u002Fwindows-app-mcp\u002FSKILL.md), including fresh-process\nprofile\u002Ftoken\u002FURL binding.\n\n## 3. Check what actually ran\n\nApply the existing\n[result rules](..\u002F..\u002F..\u002Fdocs\u002FTEST_COVERAGE.md#accepting-result-reports).\nFor inventory-bound evidence, use the\n[validator and limits](..\u002F..\u002F..\u002Fdocs\u002FPROOF_POOLS.md#validate-local-execution-evidence).\n\n- Read outputs from the exact invocation, not the newest file or exit zero alone.\n  Check executed cases, pass\u002Ffail\u002Fskip counts and failures. Label manual\n  observations as manual; do not fabricate receipts.\n- Distinguish the PR head, tested commit (including CI merge commits), uncommitted\n  source used by the run, and loaded binary\u002Fruntime.\n- After edits, commits or rebase, reassess which results and reviews still apply.\n  Source equality can establish correspondence, but cannot turn an earlier\n  review or runtime execution into a new run.\n- Retain original failures. An instrumented pass does not erase them, and\n  restoring source does not restore previously built binaries.\n- Ordinary allowed skips are not complete custom\u002Fnative proof. Identify the CI\n  run and attempt when citing results; do not present older results as newly\n  executed tests. Successful dispatch is not completed testing or review approval.\n\nMissing required review, UI or Gateway evidence stays `Not verified \u002F blocked`.\nFor an unexpected failure, preserve the attempt and identify the next bounded\ninvestigation or decision. Fix task-caused defects within scope; do not\nautomatically repair unrelated tests, rerun until green or create new PR layers.\nA failure in another task blocks this task only if this task depends on that\nresult. Otherwise continue the authorized work without borrowing its evidence.\n\n## 4. Review the intended scope\n\nAccount for both source review and the required independent rubber-duck review:\n\n- For source review, when requested or required, use the\n  [bounded local-review recipe](..\u002F..\u002F..\u002Fdocs\u002FAGENT_WORKFLOWS.md#run-a-scoped-shared-review).\n  Select mode\u002Fbase explicitly, include committed and dirty work when intended,\n  and retain Windows P0-P3 coverage. Follow its context, final-output and fallback\n  rules.\n- Before publishing non-trivial UI, MCP, node-command, setup, pairing, security,\n  permissions or diagnostics changes, obtain an independent rubber-duck review\n  of the implementation and proof plan. Follow the\n  [proof skill's review route](..\u002Fopenclaw-proof-validation\u002FSKILL.md#collect-the-selected-proof)\n  for available roles and explicit blockers. A source-review verdict alone does\n  not establish that this review was performed.\n\nReport actual scope, findings, dispositions and missing review. Self-checks are\nnot independent review, and an unavailable required review remains blocked.\nVerify findings before fixing, then refresh affected checks and review coverage.\nReviews are advisory, not runtime proof or authority to publish.\n\n## 5. Write the local body from the actual template\n\nRead the current [Windows PR template](..\u002F..\u002F..\u002F.github\u002Fpull_request_template.md).\nCopy it to a local session artifact; preserve headings, HTML comments, checklist\nitems and ordering, including security, compatibility and review conversations.\nFollow its title convention. Check only supported claims, not aspirational work.\n\nThe [proof package](..\u002Fopenclaw-proof-validation\u002FSKILL.md#pr-proof-package)\ndefines what counts as evidence. This step assembles that evidence into the\ncurrent template:\n\n- Lead with the current behavior, strongest evidence and meaningful limits.\n  Keep risks and blockers visible.\n- Put exact commands and actual outcomes\u002Fcounts in Validation. Identify the\n  tested source in Real Behavior Proof, including uncommitted work.\n- For prose-only work, label source\u002Flink checks as static assessment and runtime\n  proof as not applicable.\n- Declare every applicable pool, or use ``- `none`: \u003Creason>`` when no custom\n  pool is required.\n\nKeep supporting history in links or optional details, not an operational diary.\nCompare the completed body with the template; do not drop unanswered fields or\nclaim pending checks passed. Apply the\n[evidence\u002Fredaction boundaries](..\u002F..\u002F..\u002Fdocs\u002FPROOF_POOLS.md#evidence-and-secret-boundaries)\nand verify referenced artifacts. Private local paths are not public proof links.\n\n## 6. Hand back readiness and stop\n\nReport the exact candidate, local draft location, strongest evidence and remaining\nlimits. Say **ready for human review** only when applicable gates are satisfied.\nMissing required source or rubber-duck review means **blocked**, even when tests\npass. Name any other **blocked**, **waiting for required CI**, or **candidate\nchanged** state and the next bounded decision. A complete local draft can still\ndescribe blocked work; it is not a ready-to-merge claim.\n\nStop before publication without explicit authority. If separately authorized,\nverify the intended repository\u002Fbase\u002Fhead and live PR state, then use the host's\nsupported publication tools and their guard\u002Ffailure contracts.\nPreserve concurrent human edits; after an uncertain write, inspect the result\nbefore retrying.\nVerify published content and links without treating publication as proof.\n",{"name":5,"description":7},"main","a4dd026cc5f85a444a0fcd05eef5b86282dbda8c",".agents\u002Fskills\u002Fwindows-pr-prepare",349,"https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node","https:\u002F\u002Fopenclaw.ai","https:\u002F\u002Fgithub.com\u002Fopenclaw","Your personal, open source AI assistant.","2026-11-14 18:13:49",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add openclaw\u002Fopenclaw-windows-node --skill windows-pr-prepare",{"tool":56,"command":57},"bunx","bunx skills add openclaw\u002Fopenclaw-windows-node --skill windows-pr-prepare",{"tool":59,"command":60},"pnpm","pnpm dlx skills add openclaw\u002Fopenclaw-windows-node --skill windows-pr-prepare","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fwindows-pr-prepare\u002FSKILL.md",true,[],[65,73,94,110,129,150],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":72},"01M7416EHCYH7YCJ9XPSQ2844V","network-handoff","Use before generating, changing, or reviewing networking, IPC, or cross-process handoffs, including requests, retries, deadlines, reconnect, process launch, configuration apply, cancellation, and cleanup. Establish confirmation, identity, authority, reconciliation, and producer compatibility before implementation.","https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fnetwork-handoff\u002FSKILL.md",[],[],"2026-11-11 15:11:14",{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":36},"01M6KZCXK5PCYWWW5FCVDYPAXK","retrigger-flaky-ci","Triage failed CI on a pull request or the active branch, decide whether the failures come from its changes, and rerun only failures that are unrelated. Use when asked to retrigger, rerun, or kick CI on a PR with failing checks.","https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fretrigger-flaky-ci\u002FSKILL.md",[],[80,81,82,83,84,85,31,86,30,87,88,89,21,90,91,92,26,93],"retrigger flaky ci","ci","when","pull request","failing","rerun","triage","decide","request","active","changes","unrelated","branch","pull",{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":36},"01M698QMQN5TQES4R7ZPTYAZF5","technical-documentation","Write, review, or restructure Windows node documentation and agent instructions using current source, task-focused explanations, and explicit fact preservation.","https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Ftechnical-documentation\u002FSKILL.md",[],[101,102,103,32,104,35,105,106,19,107,108,109],"technical","documentation","technical documentation","instructions","write","task","agent instructions","preservation","source",{"skillNo":111,"name":112,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":114,"categories":115,"keywords":116,"publishedAt":36},"01M698QKPXJ8ZG34MYHPPCAT4T","pr-simplify","Simplify your own Windows-node change to its smallest complete fix while preserving distinct regressions, platform contracts, and required proof. Use after implementation before PR preparation, after large repairs, or when asked to shrink a change or remove slop.","https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fpr-simplify\u002FSKILL.md",[],[117,21,118,119,120,121,122,19,123,82,31,124,125,126,127,128,32],"pr simplify","simplify","preparation","contracts","proof","implementation","change","fix","complete","remove","slop","platform",{"skillNo":130,"name":131,"nameZh":6,"description":132,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":140,"publishedAt":36},"01M698QJEA384YPM8R9CPZ1YFD","openclaw-testing","Choose proportional OpenClaw Windows node tests, builds, packaging checks, and runtime proof, then diagnose failures without weakening required gates.","https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fopenclaw-testing\u002FSKILL.md",[135],{"categoryNo":136,"categoryName":137,"taxonomyType":138,"taxonomyTypeDesc":139},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",[9,141,142,143,19,144,121,32,145,146,147,148,149],"openclaw testing","testing","diagnose","runtime","gates","runtime proof","tests","choose","packaging",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":159,"publishedAt":36},"01M698QH7JVVA09769ER9P2TKY","openclaw-proof-validation","Plan and collect OpenClaw Windows validation\u002Fproof: tests, rubber-duck review, UI evidence, MCP output, and gateway runtime proof.","https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw-windows-node\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fopenclaw-proof-validation\u002FSKILL.md",[156],{"categoryNo":157,"categoryName":158,"taxonomyType":138,"taxonomyTypeDesc":139},"01JBSEEDCAT000000000000023","Skill与智能体开发",[121,34,160,9,146,161,162,163,164,165,35,147,166,144,29,19,167,168],"openclaw proof validation","duck","ui","ui evidence","plan","output","collect","gateway","MCP"]